1. Bottom Line & Directional Bias
Call: Bearish NG=F. The prior-session settle of 3.04 (2026-10-02) is the reference; the view is invalidated on a settle above 3.14 (pivot R2). Three reasons underpin the call. First, price is in the lower quartile of the 20-day 2.9–3.4 channel (position 27%) and below pivot R1 at 3.09, with the 5D change at -5.89% (settle) — momentum is down, not consolidating. Second, the curve is in contango (M1-M2 -0.327, -10.03%), which is the opposite of prompt tightness; there is no scarcity signal in the front of the curve to force a squeeze. Third, managed-money positioning remains net short at -3.57% of open interest (as of 2026-09-22) with a crowding percentile of 27 — the short is not crowded, so there is room for further selling without a violent short-covering reversal. The main counterweight is seasonality: the next 20 sessions have averaged +4.38% over the last 15 years (median +0.96%, up 9 of 15). That is a real risk, but it is a small sample and the current contango structure does not support a sustained seasonal bid. Invalidation: a settle above 3.14 (R2) would break the lower-high sequence and force a reassessment.
2. Price Action & Technical Analysis
The prior-session settle was 3.04 (2026-10-02), with a 1D change of +2.29% (settle) — a bounce within a downtrend, not a reversal. The 5D change is -5.89% (settle), and the 20D change is -0.07% (settle), meaning the market has gone nowhere over a month but has lost ground sharply over the past week. The 20-day channel is 2.9–3.4, and the settle sits at position 27% of that range — lower quartile. The 52-week range is 2.48–7.83, so the market is in the bottom third of its annual range, but well above the 52-week low.
ATR14 is 0.127, which is 4.19% of price — a full daily range, not a one-sided move. RV20 is 43.2%, which is elevated in absolute terms and consistent with the recent 5D decline. The pivot structure is: P 3, R1 3.09, S1 2.95, R2 3.14, S2 2.86. The settle of 3.04 is above P (3) but below R1 (3.09), so the immediate bias is neutral-to-weak within the pivot grid; a failure at R1 keeps the short case intact, while a break below S1 (2.95) opens S2 (2.86).
In early Asian trade on the report date, the market is trading around the 3.04 area; this is an unfinished bar and should not be treated as a settle. The last completed weekly bar (2026-09-21–2026-09-25) had an open of 3.04, high of 3.4, low of 2.98, and close of 3.23, a +5.98% w/w gain. The current week (from 2026-09-28, five sessions) is not closed; the last price of 3.04 (-5.89%) is an unfinished-week figure and cannot be used for weekly-close conclusions. The key technical takeaway: the market failed to hold the prior week's gains and is back at the prior week's open, which is a bearish retracement. A settle below 2.95 (S1) would confirm the next leg lower; a settle above 3.14 (R2) would negate the bearish structure.
3. Supply-Demand Balance & Fundamental Drivers
The curve is in contango: M1-M2 is -0.327, or -10.03%, with a roll yield of -120.41% and a slope of -0.0103. Contango means the front month is cheaper than the second month, which is the market's way of paying for storage — it signals ample near-term supply relative to demand. This is the single most important fundamental signal in the data: there is no prompt tightness. For a long, contango is a roll cost; for a short, it is a tailwind, because the passage of time works against holders of the front month.
The COT data (as of 2026-09-22) shows open interest of 1,837,146 contracts, with managed-money longs at 266,563 and shorts at 332,110, for a net of -65,547. That is a net short position, and the week-over-week change was +34,658 — meaning shorts were reduced (net short got smaller) in the latest reporting week. The prior week (2026-09-15) had a net of -100,205, and the week before that (2026-09-08) was -96,742. So the trend over the last three reports is a reduction in net short exposure, which is a mild positive for price, but the absolute level is still net short. The crowding percentile is 27 (3-year percentile of netPct), which is not extreme — the short is not crowded. The CTA proxy is 74, which is high, suggesting trend-following flows are positioned for the move; hedge percentage is 27.4%.
On the macro side, the US 10-year yield is 5.277 (up 0.76%), and DXY is 101.92 (down 0.17%). A lower dollar is typically a mild tailwind for dollar-denominated commodities, but the move is small and the yield is high — the macro transmission to natural gas is weak. The more relevant macro driver is the FOMC minutes on 2026-10-07 (ET 14:00), which could move rates and the dollar, but the direct impact on gas is second-order. The EIA crude and gasoline stock changes on 2026-10-07 (ET 10:30) are for crude and gasoline, not natural gas, but they can influence the broader energy complex. The ISM Services PMI on 2026-10-05 (ET 10:00) is a demand-side indicator for the US economy; a strong print could support industrial demand expectations, but the surprise threshold is outside F±1.4 (forecast 54, previous 55.4).
Net: the fundamental structure — contango, net short positioning, and no prompt tightness — supports the bearish call. The main fundamental risk is a weather-driven demand spike or a supply disruption, neither of which is in the data.
4. Positioning & Fund Flows
The latest CFTC report is as of 2026-09-22, which is 11 days old — it is not the current week's positioning. Managed-money net was -65,547 contracts, or -3.57% of open interest. The crowding percentile is 27, which is low on the 3-year window; this is not a crowded short. The week-over-week change was +34,658, meaning shorts covered (net short decreased) in that reporting week. The prior two weeks saw net shorts of -100,205 and -96,742, so the trend is toward less short exposure. This is a mild divergence from the price action: price fell 5.89% over the last 5 days (settle), but the most recent COT data (which predates that move) showed shorts reducing. The resolution is that the COT data is stale; the price decline since 2026-09-22 likely reflects new short selling that is not yet in the report. If the next COT report shows a re-increase in net shorts, it would confirm the bearish trend; if it shows further short covering, it would be a warning.
RV20 is 43.2%, which is high. The ^OVX (WTI implied vol) is 51, at the 49th percentile of its 1-year range; ^GVZ (gold implied vol) is 23.23, at the 15th percentile; ^VXSLV (silver implied vol) is 36.9; ^VIX is 15.31, at the 15th percentile. The energy complex (via OVX) is pricing moderate event risk, while equity and gold vol are low. For natural gas, the absence of a direct IV reading means we rely on RV20 as the volatility gauge; at 43.2%, the market is moving, and options on gas are likely expensive relative to other assets, but we cannot make a precise IV-vs-RV call. The positioning takeaway: the short is not crowded, so there is room for further downside without a squeeze; the main risk is a short-covering rally if the next COT shows a sharp reduction in net shorts.
5. Cross-Asset Relative Value
The relevant comparisons are: DXY at 101.92 (-0.17%) and ^TNX at 5.277 (+0.76%). A rising 10-year yield and a slightly weaker dollar are a mixed macro backdrop for commodities. The energy complex is represented by ^OVX at 51 (49th percentile), which is mid-range — not signaling extreme fear or complacency. The natural gas market is not directly linked to these ratios in the data, so the cross-asset view is: the macro backdrop is not a strong driver for gas right now, and the contango structure is the dominant relative-value signal. The view: cross-asset signals are neutral-to-slightly-bearish for gas, with the contango being the key differentiator.
6. Historical & Seasonal Patterns
The seasonality block shows that for the same calendar start, over the next 20 sessions, the last 15 years have averaged +4.38%, with a median of +0.96%, and the market was up in 9 of 15 years. The best year was 2025 (+24.07%) and the worst was 2021 (-10.06%). This is a small sample and should be treated as context, not a forecast. The positive mean is skewed by the +24.07% outlier; the median of +0.96% is much more modest. The hit rate of 9/15 (60%) is above a coin flip but not overwhelming. For a bearish call, this seasonality is the main risk: if the market follows the average, it would rally into late October. However, the current contango structure and the 5D decline of -5.89% suggest that this year may not follow the seasonal pattern. The view: seasonality is a headwind for the short, but the small sample and the current fundamental structure (contango, net short positioning) argue against a strong seasonal bid. The bearish call remains, but with awareness that a seasonal rally is the primary risk.
7. Scenario Analysis (Base / Bull / Bear)
Base case (55% probability): Bearish continuation. Trigger: price fails to hold above 3.09 (R1) and settles below 3 (P). Target: 2.95 (S1) initially, then 2.86 (S2). Action: maintain short exposure, add on rallies to 3.09–3.14 with a stop above 3.14. This scenario is consistent with the call in Section 1. The rationale: contango, net short positioning, and the lower-quartile position in the 20-day channel all point to further downside. The 5D change of -5.89% (settle) shows momentum is down, and the bounce of +2.29% (settle) on the prior session is likely a dead-cat bounce within the downtrend.
Bull case (25% probability): Seasonal rally / short squeeze. Trigger: a settle above 3.14 (R2) on strong volume, or a bullish EIA report (not in the calendar for gas, but the crude/gasoline data on 2026-10-07 could spill over). Target: 3.23 (prior completed weekly close), then 3.4 (prior weekly high). Action: if the market settles above 3.14, the bearish call is invalidated; stand aside and reassess. The bull case is supported by the positive seasonality (mean +4.38%, median +0.96%) and the fact that the short is not crowded (crowding percentile 27), which means a short-covering rally could be sharp. The risk is that the market has already priced in a lot of bearish news, and any supply disruption or cold snap could trigger a squeeze.
Bear case (20% probability): Accelerated decline. Trigger: a settle below 2.95 (S1) with rising volume, or a bearish macro surprise (e.g., weak ISM Services PMI on 2026-10-05, or hawkish FOMC minutes on 2026-10-07). Target: 2.86 (S2), then 2.7 (psychological). Action: add to shorts on a break below 2.95, with a stop at 3.09. The bear case is supported by the contango structure and the fact that the 52-week low is 2.48, leaving room to fall. The risk is that the market is already oversold on a 5D basis (-5.89%), so a bounce could precede further downside.
8. Trading Strategies & Risk Management
Strategy 1: Short on rallies. Entry: 3.09–3.14 (R1 to R2 zone). Stop: 3.2 (above R2, approximately one ATR of 0.127 from entry). Target: 2.95 (S1) then 2.86 (S2). Horizon: 1–5 days. Size: 1.5% risk per trade. Conviction: 7/10. This strategy aligns with the bearish call and uses the pivot structure to define risk. The stop at 3.2 is beyond R2 (3.14) and more than one ATR away from the entry midpoint, so it is not inside normal daily noise.
Strategy 2: Momentum short on a break below S1. Entry: on a settle below 2.95 (S1). Stop: 3.09 (R1). Target: 2.86 (S2) then 2.75. Horizon: 1–5 days. Size: 1% risk per trade. Conviction: 6/10. This is a trend-following add-on to the core short. The risk is a false breakdown; the stop at 3.09 is above the breakdown level and approximately one ATR from the entry.
Both strategies are in the direction of the call. If the market settles above 3.14 (R2), the bearish call is invalidated and both strategies should be closed. Do not add to shorts if the market is above 3.14.
9. This Week's Data Calendar
- BJT 10-05 22:00 | ET 10-05 10:00 — ISM Services PMI SEP (F:54, P:55.4; surprise if outside F±1.4). Impacts DXY, GC, SI; second-order for NG.
- BJT 10-07 04:30 | ET 10-06 16:30 — API Crude Oil Stock Change OCT/02. Impacts CL, BZ; spillover to energy complex.
- BJT 10-07 22:30 | ET 10-07 10:30 — EIA Crude Oil Stocks Change OCT/02. Impacts CL, BZ.
- BJT 10-07 22:30 | ET 10-07 10:30 — EIA Gasoline Stocks Change OCT/02. Impacts CL, BZ.
- BJT 10-08 02:00 | ET 10-07 14:00 — FOMC Minutes. Impacts GC, SI, DXY; macro risk for all assets.
This report is generated automatically from public quantitative and macro data for research and market tracking only. It does not constitute investment advice or a recommendation to trade.